The outcome of many genetic disorders is influenced by modifying genes. The long term goal of these studies is to identify mechanisms for modulating mutations that affect brain and behavior. Such epistatic interactions may be specific for a single mutation, applicable across genes in a biological pathway, or applicable across a class of mutagenic events. From studies on the mouse neurodegeneration mutant vibrator, we identified a genetic modifier, Mvb1, that acts on mutations caused by endogenous retroviruses. Retroviruses account for ~15% of spontaneous mouse mutations. Suppressing alleles of Mvb1 increase expression of normal gene products derived from mutant genes, including vibrator and a model of branchiootorenal dysplasia, Eya1, in which insertion into an intron has reduced gene expression. We now have complementation evidence that Mvb1 is a naturally occurring allelic difference in an mRNA export factor, Nxf1. Two amino acid substitutions that distinguish suppressing and non-suppressing alleles. This factor is recruited to spliced mRNA, but also directly binds RNA elements in unspliced retroviral genomes, in L1 LINEs, and potentially some non-repetitive cellular RNAs. Retrotranspositions, particularly L1 LINES, have been identified in several inherited human diseases. This proposal will impact human health through improved understanding of retroviral control, retroelement impact on host gene expression and creation of new classes of mouse models. This renewal has four specific aims. (1) Test the range of Mvb1-mediated genetic suppression in vivo by genetic crosses to select retroviral mutations and in cell culture by manipulation of reporter constructs. (2) Test competing hypotheses for the functional difference between allele using amino acid replacement alleles and quantitative measurements of key protein-protein interactions. (3) Test the modulation of heterologous target gene expression by Mvb1 in cell culture and transgenic mice. (4) Test the hypothesis that suppressing Mvb1 alleles arose through recent positive selection will be tested by comparing nucleotide diversity and extended haplotype homozygosity at Mvb1 compared to other loci.